ABSTRACT Not only do children from disadvantaged families have significantly worse health at birth and poorer adult outcomes, but the effects of poor health at birth appear to extend across generations. Given these patterns, it has been posited that interventions that improve early life health could yield high returns and help to break the intergenerational cycle of disadvantage. Despite the clear public health importance of this topic, little research exists to help us understand the long- reaching effects of health at birth and the role for early intervention in the United States. This project proposes to provide new evidence on this front using a novel dataset that links birth records for all individuals born in California between 1960 and 2014 to large-scale, federal survey and administrative data with information on later life mortality, disability, educational attainment, earnings and income, and reliance on public assistance. To our knowledge, this will be the first such dataset in the U.S. and will be unprecedented both in terms of its coverage and scope. Using this new dataset, we will estimate the long-term effects of health at birth on adult disability, mortality, and economic outcomes by comparing outcomes among twins born with different birth weights (Aim 1). This estimation strategy adjusts for any unobserved family-specific characteristics that might confound estimates of the causal impact of health at birth on outcomes later in life. Using a similar identification technique, we propose the first analysis of the intergenerational effects of health at birth on long-term health and economic outcomes (Aim 2). We will estimate the impact of a mother's birth weight on the outcomes of her children, both at the time of birth and later in life. This will provide new information on the effects on her child's mortality risk, disability, and achievement throughout life. We will also explore the different mechanisms driving the intergenerational effects of mother's health at birth. For both Aims 1 and 2, we will obtain estimates for the overall population and for subgroups defined by race, ethnicity, and socioeconomic status. The scope and size of our dataset will allow us to generate precise estimates across these subgroups and will provide important new information on the role of early life health in explaining persistent health and socioeconomic disparities for certain disadvantaged groups. Finally, we will examine the potential role for early intervention in altering health at birth and long-term outcomes using a case study of a large-scale, enhanced prenatal care program. We propose to evaluate the short- and long-term effects of a statewide intervention targeted to low-income pregnant women in California for the children who benefit from this intervention while in utero (Aim 3). To identify the effects of this intervention, we will use a natural experiment approach that takes advantage of the program's staggered implementation across counties and the resulting variation in exposure to the program. This will be the first rigorous evaluation of this large-scale, publicly-funded program and, to our knowledge, the only evaluation of the long-term effects of an enhanced prenatal care program, despite the widespread use of these programs as a policy tool to promote infant health among disadvantaged populations. The analyses proposed under this study will generate new and actionable information for policymakers aiming to better understand and mitigate the long- term impacts of inequalities in early life health.